Generally, in high definition image processing, image data corresponding to an image is input via a scanner. Then, the image data is typically transmitted after enhancing the data relating to an edge portion of the image so that the image can be transmitted with a high definition. For example, when text is scanned via a facsimile machine, the edges of various letters may be enhanced so that the resolution and definition of the transmitted letters is high.
In order to process an image to enhance its edges, the data of an A4 size input via a scanner is stored in 1728 facsimile data buffers and processed by software. Specifically, the current pixel value of the data stored in the data buffer is compared with adjacent pixel values before determining whether the current pixel value should be black or white. As a result, the dark portions of the corresponding text become even darker and the bright portions become even brighter, thereby producing the high definition of the text.
When an image is processed in one dimension, the pixel values as shown in FIG. 1 are processed to determine if the pixels should be black or white. In particular, as shown in FIG. 2, the current pixel value is weighted by a factor of +2, and the pixel values to the left and right of the current pixel value are weighted by a factor of -1. Then, the current pixel value is enhanced as follows. First, the weighted values of the left pixel, the right pixel, and the current pixel are combined to produce a combined weighted value. Then, the weighted value is divided by the magnitude of the combined weighting factors (e.g. .vertline.-1.vertline.+.vertline.+2.vertline.+.vertline.-1.vertline.=4), and the combined weighted value is divided by such magnitude to produce a calculation value. Finally, the current pixel value is enhanced by adding the calculation value to the current pixel value.
For example, FIG. 1 is a diagram of pixels [1], [2], [3], . . . [1728], and the pixels [1], [2], [3], [4], [5], and [6] respectively have pixel values which equal 10, 10, 5, 5, 5, and 10. Accordingly, if pixel [2] is the current pixel and is processed according the method shown in FIG. 2, the pixel value 10 of the pixel [2] is enhanced to produce the enhanced pixel value 11.25. Specifically, the current pixel value 10 of the pixel [2] is weighted by a factor of +2 to produce a weighted current pixel value of 20, the left pixel value 10 of the pixel [1] is weighted by a factor of -1 to produce a weighted left pixel value -10, and the right pixel value 5 of the pixel [3] is weighted by a factor of -1 to produce a weighted right pixel value -5. Then, the weighted current, left, and right pixel values are combined to produce a combined weighted pixel value +5 (i.e. -10+20-5), and the combined weighted pixel value is divided by the magnitude of the combined weighting factors (i.e. 4) to produce a calculation value 1.25. Then, the calculation value is added to the current pixel value 10 of the pixel [2] to produce the enhanced pixel value 11.25.
The enhanced pixel values of pixels [2] to [5] are illustrated in FIG. 3. Since the enhanced pixel values of pixels [2], [3], and [5] are different than the corresponding original pixel values, an image having a higher definition can be obtained.
When the image is processed in one dimension in the above manner, the image can be processed without major problems since only the current, left, and right pixel values need to be stored and processed via software. However, when an image is processed in two dimensions, the image cannot be efficiently processed. For example, as shown in FIG. 4, a current pixel is two dimensionally processed by weighting its pixel value by a factor of 8 and by weighting the eight pixels surrounding the current pixel by a factor of -1. Then, the nine weighted values are processed to produce the enhanced pixel value for the current pixel. Since nine pixels are needed to two-dimensionally process the current pixel value, 1728.times.2 buffers are required for the processing. The reason for the large number of buffers is that all the pixel values with respect to the previous line and the line before the previous line need to be stored. Accordingly, the two dimensional processing method is disadvantageous because a lot of buffers are needed for software processing and the calculation time increases. Thus, the processing speed is dramatically reduced.